Critical Care Review
Paul Wischmeyer M.D.
Associate Professor of Anesthesiology
University of Colorado Health Sciences Center
Case Study
 It’s now 3 pm and you get the dreaded call there
is an emergency case for an exploratory lap that
needs to go now !
Case Study
 58 y.o. male with firm, distended abdomen and severe
metabolic acidosis (lactate- 5.0) coming to OR for
exploratory laparotomy
 Vent Settings: Pressure Control - 34 Peep- 12 R- 24,
FiO2= 100%, PiP- 46
 ABG: 7.20 / pCO2 - 50 / pO2 - 65
 CXR- reveals 3 quadrant infiltrates
 HR 110, BP 75/40, C.I. - 4.0 SVR- 458
 Drips: vasopressin and insulin
 U/O - 5 cc last hour
What can we do to save this
patient today… that perhaps
we could not have a few
years ago ??
But first… is he “critically ill” ?
What makes for a “critically ill”
patient ?
Parasite
Virus
Infection
Fungus
Severe
Sepsis
shock
Sepsis
SIRS
Severe
SIRS Trauma
Bacteria
BSI
Adapted from SCCM/ACCP Consensus Guidelines
Burns
The Sepsis Continuum
SIRS
Sepsis
Severe Sepsis
SIRS due to
infection
2 of the following:
Organ Dysfunction:
• T >38oC or <36oC
• Shock/hypotension
• HR >90
• RR >20
• WBC >12,000 or <4,000
or >10% bands
• Acute lung injury
• Renal
• Hepatic
• Hematologic
• CNS
• Metabolic acidosis
Bone RC, et al. Chest 1992;101:1644-55.
DEFINITIONS


MOFS: multiple organ failure syndrome
– Progressive, persistent, hyperdynamic,
hypermetabolic state associated with
gradual deterioration of multiple organs,
lungs usually being the first
MODS: multiple organ dysfunction
syndrome
– Organs rarely fail abruptly
PATHOPHYSIOLOGY


The (abnormal?) host response to insult
Prolonged shock state
– Can be from any form of shock (hypovolemic,
cardiogenic, distributive/vasodilatory, or obstructive)
– Cellular level: mitochondrial dysfunction and
“cytopathic hypoxia”
– Mitochondria “leak” free radicals, and activate nuclear
factor kappa B (NF-kB), a transcription factor for many
pro-inflammatory mediators
– Mediators may stimulate cytokines (TNF, interleukins),
complement system, coagulation and fibrinolytic
system, and cellular system (macrophages,
neutrophils, endothelial cells, and platelets)
PATHOPHYSIOLOGY


“2nd hit” phenomenon
– The initial insult “primes” inflammatory
system, and a second insult amplifies the
response
“Gut” hypothesis
– GI tract is the “undrained abscess” causing
MODS. The gut leaks bacteria/products
– Possibly via lymphatics (ALI/ARDS)
– GI tract is largest immune organ (GALT)
CLINICAL MANIFESTATIONS
of MOFS

Patients appear to stabilize after
resuscitation to a “hypermetabolic” state

Lungs usually the first (ALI/ARDS) to injure

Sequence of other organs influenced by comorbidities

Renal dysfunction usually follows second

Liver failure late

Usually die from hypotension unresponsive
to pressors
TREATMENT
 THERE
IS NO KNOWN
TREATMENT!!!!
PREVENTION



Best defined by literature that shows
improved outcomes or survival in patients
typically at risk for MODS
Prospective, randomized, placebo
controlled trials (PRCT’s)
Grade A or B recommendations
– A: 2 separate Level I trials
– B: only 1 Level I trial
Once we have a critically ill
patient… What does the
latest literature say…
1) Does Not Work
2) May Work
3) Does Work
What Doesn’t Work
Colloid vs Crystalloid Controversy
0.01
Lowe
Lucas
Butros
Virgillo
Moss
Goodwin
Modic
Rackow
Shires
Metildi
Sade
Karanko
Davidson
London
Pocka
Overall
1977
1978
1979
1979
1981
1983
1983
1983
1983
1984
1985
1987
1991
1992
1994
0.1
1
10
0.68
0.07
2.22
1.07
2.43
0.27
1.08
1.23
1.0
0.82
1.83
2.37
1.0
0.22
0.90
0.97
Choi PTL, et al. Critical Care Med 1999;27:200-10.
Schierhout G, et al. BMJ 1998;316:961-4.
Favors
Crystalloid
Favors
Colloid
100
Dopamine for Renal Dysfunction

RCT, 328 patients

Included: 2 SIRS criteria plus oliguria or
creatinine >1.7 or 24-hour rise in creatinine >0.9


Dosing: Dopamine 2 g/kg/min via CVC
No differences found in
– Peak creatinine, fraction with creatinine >3.4
– Percent requiring renal replacement therapy
– Time to recover renal function
– Furosemide dose, ICU LOS, hospital LOS
– Time on ventilator, survival
Bellomo R, et al. Lancet 2000;356:2139-43.
Supranormal Oxygen Delivery
0.1
Author
Shoemaker
Tuchsmidt
Yu
Boyd
Hayes
Yu
Gattinoni
Overall
1988
1992
1993
1993
1994
1994
1995
1
10
RR
0.10
0.69
1.00
0.25
1.60
0.93
0.98
0.86
Favors high DO2
Favors normal DO2
What Might Work
Early Goal-directed Therapy
for Septic Shock



RCT , n = 263
Septic shock unresp to 20 ml/kg crystalloid or
lactate > 4
Rx (all patients receive CVP and SvO2 monitor
– Traditional: CVP 8-12, Vasopressor for SBP <
90 mm Hg, keep UOP > 0.5 ml/kg/hr
– Investigation: As above + RBCs for hct < 30
AND SvO2 < 70, if fails add dobutamine to
dose 20 ug/kg/min
Rivers et al NEJM 345:1368 2001
Early Goal-Directed Therapy
Patient Randomized
Early GoalDirected Therapy
Standard
Therapy
CVP > 8-12 mm Hg
CVP > 8-12 mm Hg
MAP > 65 mm Hg
MAP > 65 mm Hg
Urine Output > 0.5 ml/kg/hr
Urine Output > 0.5 ml/kg/hr
ScvO2 > 70%
SaO2 > 93%
Antibiotics given at discretion of
Hct > 30%
treating clinicians
At least 6 hours
of EGDT
Rivers E. N Engl J Med 2001;345:1368-77.
Transfer to ICU
ICU MDs blinded to
study treatment
As soon as
possible
EGDT in Septic Shock:
Treatments actually received (0-6 hrs)
Traditional




Fluids (mL)
RBCs (%patients)
Vasopressor (%pts)
Dobutamine (%pts)
3500
19
30
1
Rivers et al NEJM 345:1368 2001
EGDT
5000
64
27
14
EGDT - Outcome
50
45
40
35
28-day Mortality
30
25
P = 0.01*
20
15
10
5
0
Traditional
EGDT
*Key difference was in sudden cv collapse not MSOF
Rivers et al NEJM 345:1368 2001
Vasopressin…Why
Vasopressin ??
BP and Vasopressin Levels
After AVP for Septic Shock
SBP
(mm Hg)
or
AVP level (pg/mL)
160
120
80
SBP
40
0
Baseline
Landry DW, et al. Circulation 1997;95:1122-5.
Landry DW, et al. Crit Care Med 1997;25:1279-82.
AVP 0.04
U/min
AVP Off
AVP 0.01
U/min
Vasopressin Summary

No large, randomized controlled trials

Raises MAP by increasing SVR in
vasodilatory shock

If effective, acts rapidly and at low doses

Probably synergistic with catecholamines

At this time, no data suggest
improved outcomes
Intensive Insulin Therapy in
Critically Ill Patients
Van den Berghe G, et al. N Eng J Med 2001;345:1359-1367
Intensive Glucose Control
in the Critically ILL





RCT, n = 1548
Mechanically ventilated SICU patients
Treatments
– Titrate blood glucose 80-110
• VS
– Titrate blood glucose 180-200
All patients received 200-300 gms glucose/d on
day - 1 (?D10W)
TPN w/in 24 h of adm (60-80% as glucose cals)
Berghe et al NEJM 345:1359 2001
Intensive Glucose Control
Treatments received





Conventional
Patients on insulin
39%
Insulin (median u/d)
33
Duration (%ICU days)
67
AM glucose (all pts)
153
AM glucose (insulin pts) 173
Van Den Berghe et al NEJM 345:1359 2001
Intensive
99%
71
100
103
103
Intensive Glucose Control
Outcome
In-hospital
mortality
30
26.3
25
20
All patients n = 1548
ICU > 5d n =451
16.8
15
10.9
10
7.2
5
P = .01
both comparisions
0
Conventional
Berghe et al NEJM 345:1359 2001
Intensive
Intensive Insulin Therapy in Critically
Ill Patients: Kaplan-Meier Curves
Van den Berghe G, et al. N Eng J Med 2001;345:1363
Intensive Insulin Therapy in Critically
Ill Patients: Morbidity
Percent of Patients Requiring >14 Days of Ventilatory Support
Percent of patients
20%
p = 0.003
15%
11.9%
10%
7.5%
5%
n=783
n=765
0%
Conventional Intensive
Van den Berghe G, et al. N Eng J Med 2001;345:1365
Intensive Insulin Therapy in Critically
Ill Patients: Morbidity

Percent of Patients with Renal Impairment
Conventional
n=783
Intensive
n=765
P-value
Peak plasma
creatinine >2.5
mg/dL
12.3%
9.0 %
0.04
Peak plasma
urea nitrogen
>54 mg/dL
11.2 %
7.7 %
0.02
Dialysis or
CVVH
8.2 %
4.8 %
0.007
Van den Berghe G, et al. N Eng J Med 2001;345:1359-1367
Intensive Insulin Therapy in Critically
Ill Patients: Morbidity

Percent of Patients with Bloodstream Infections
Conventional
n=783
Intensive
n=765
P-value
Septicemia
during
intensive care
7.8%
4.2%
0.003
Treatment with
antibiotics > 10
days
17.1%
11.2%
<0.001
Van den Berghe G, et al. N Eng J Med 2001;345:1365
Intensive Insulin Therapy in Critically
Ill Patients: Author’s Conclusion
“Intensive insulin therapy to
maintain blood glucose at or below
110 mg per deciliter reduces
morbidity and mortality among
critically ill patients in the surgical
intensive care unit.”
Van den Berghe G, et al. N Eng J Med 2001;345:1359
with Adrenal Insufficiency:
Study Design
TIME 0
Onset of shock
Eligibility
and
ACTH test
Randomization
At 8 hours
Hydrocortisone IV 50 mg
QID
+
9a Fludrocortisone PO
50 µg/d for 7 days
Annane D. Crit Care Med 2000;28(suppl): Abstract 63.
PLACEBO
For 7 days
with Adrenal Insufficiency:
Results
28-Day Cumulative
Survival in All Patients
Cumulative Survival Rate
28-Day Cumulative Survival
in Non-Responders
1.00
1.00
0.80
0.80
TREATMENT
TREATMENT
0.60
0.60
0.40
0.40
PLACEBO
0.20
PLACEBO
0.20
P = 0.02
P = 0.01
0.00
0.00
0
7
14
21
28
Time (days)
Annane D. Crit Care Med 2000;28(suppl): Abstract 63.
0
7
14
Time (days)
21
28
What Works
Balancing Ventilation Priorities
Inadequate
Tidal Volume
or PEEP
Excessive
Tidal Volume
or excessive
PEEP
Acute Lung Injury
Before Ventilation
Consequences:
• Atelectasis
• Hypoxemia
• Hypercapnia
TNF
IL-6
Consequences:
• V/Q mismatch
• Alveolar-capillary injury
• Inflammation
• Pulmonary hypertension
• “Barotrauma”
Ventilator Management




Assist control mode
Reduce TV to 6 mL/kg predicted body
weight
Keep plateau airway pressure <30 cm H2O
Maintain SaO2 / SpO2 88%-95% using this
scale:
FiO2
.3 .4 .4 .5 .5 .6 .7 .7 .7 .8 .9 .9 .9 1.0
PEEP 5
5
8
8
10 10 10 12 14 14 14 16 18 20-24
The Acute Respiratory Distress Syndrome Network. N Engl J Med 2000;342:1301-8.
Ventilator Management

Accept mild respiratory acidosis
– If pH <7.30 increase rate (max 35)
– If acidosis persists and rate = 35,
consider NaHCO3
– If acidosis refractory/unresponsive, may
raise TV to achieve pH >7.15

Perform a spontaneous breathing trial
daily if
– Shock absent
– Spontaneous efforts present
– FiO2 0.4 and PEEP = 8
The Acute Respiratory Distress Syndrome Network. N Engl J Med 2000;342:1301-8.
Mortality Prior to Discharge
40
Mortality (%)
P=0.0054
30
20
10
0
6 mL/kg
12 mL/kg
The Acute Respiratory Distress Syndrome Network. N Engl J Med 2000;342:1301-8.
Spontaneous Breathing Trials

Respiratory therapy driven protocol

Daily screening of all ventilated patients:
– No shock
– FIO2 0.4, PEEP 8
– Awake, not-paralyzed

Two hour spontaneous breathing trial

Extubation decision by primary team
Ely EW, et al. N Engl J Med 1996;335:1864-9.
Patients on Ventilator (%)
Value of Spontaneous
Breathing Trials
100
80
Uncontrolled
Protocol
60
40
P<0.01
20
0
0
3
6
9
12 15 18 21 24 27 30
Days
Ely EW, et al. N Engl J Med 1996;335:1864-9.
Sepsis Specific Therapy
At Last a Reality
The PROWESS Trial:
Drotrecogin Alfa (Activated)
in Patients with Severe Sepsis

Anticoagulant
– Inactivates coagulation factors Va, VIIIa
– Inhibits formation of thrombin

Pro-fibrinolytic
– Allows activity of tissue plasminogen
activator (endogenous TPA)

Antiinflammatory
– Reduces IL-6 and
proinflammatory cytokines
Drotrecogin Alfa (Activated) in
Severe Sepsis: Phase III Study



Randomized 1:1
Blinded
Large N=1690




Placebo-controlled
164 centers
11 countries
Severe sepsis
Infection + 3 SIRS criteria
and acute (<24 hr) organ failure
enrollment in
<48 hours
Day 28
Outcome
96 hr study
drug infusion
Routine Care
Bernard GR, et al. N Engl J Med 2001;344:699-709.
28-Day Survival All-cause Mortality
Percent Survivors
100
90
Drotrecogin alfa (activated)
(N=850)
80
Placebo
(N=840)
70
P=0.006 (stratified log-rank test)
0
0
7
14
21
Days from Start of Infusion
28
Mortality by Site of Infection
Drotrecogin
Alfa
N (activated) Placebo
Overall
Lung
906
25.0%
33.6%
Intra-Abdominal
337
27.7%
30.5%
Other
276
22.3%
28.5%
Urinary Tract
171
21.2%
20.9%
-50
-40 -30
-20
-10
0
10
Relative Risk (%)
20
30
Other Important New Data
You Should Be Aware Of !!
“Appropriate” antibiotics
Kollef, Chest 1999
 Inadequate antimicrobial treatment of infection
 Defined as microbiologic documentation of infection (ie,
positive culture result) not being effectively treated at time
of identification
 Absence of antimicrobial agents directed at specific class
of microorganisms (absence of tx for fungemia due to
Candida) and administration of agent to which
microorganism responsible for infection were resistant
(eg, empiric tx with methacillen for pneumonia
subsequently attributed to methacillen-resistant
Staphylococcus aureus [MRSA] based on culture results).
“Appropriate” antibiotics
Kollef, Chest 1999
169 patients out of 2000 surveyed were “inappropriately
treated” at diagnosis of infection
Mortality rate in this group was 52.1 %
12.2% mortality in appropriately treated group versus
52.1% (NNT = 2.5)
“Invasive” strategy for VAP diagnosis
Fagon, Ann Int Med 2000
Examined diagnosis of ventilator
associated pneumonia via bronchoscopic
BAL samples and their quantitative
cultures
Versus noninvasive isolation of
microorganisms by nonquantitative
analysis of endotracheal aspirates, and
clinical practice guidelines.
“Invasive” strategy for VAP diagnosis
Fagon, Ann Int Med 2000
Patients who recv’d invasive diagnosis had:
Reduced mortality at day 14 (16.2% vs. 25.8%; p <
0.02)
Decreased Sepsis-related Organ Failure
Assessment scores at day 3 and day 7
Decreased antibiotic use (mean number of
antibiotic-free days, 5.0+/-5.1 and 2.2+/-3.5; P <
0.001).
“Invasive” strategy for VAP diagnosis
Fagon, Ann Int Med 2000
 Conclusions“Compared with noninvasive
management strategy, invasive
management strategy was significantly
associated with fewer deaths at 14 days,
earlier attenuation of organ dysfunction,
and less antibiotic use in patients
suspected of having ventilatorassociated pneumonia.”
Intensivist led multidisciplinary
ICU team
Young, Effective Clinical Practice
2000
Concept of closed ICU service led
by ICU physician
Up to 60% reduction in mortality
Optimal Hemoglobin in the Critically
Ill Patient !?!
Clearly higher Hgb achieved via
transfusion is not helpful and may be
harmful
Is there a lower threshold?
Transfusion Requirements in
Critical Care
Multicenter, RCT
Subjects
– Acutely ill in ICU, Hgb < 9.0
– Excluded if: chronic anemia, ongoing
bleeding, admission after CABG
Hebert et al. NEJM 1999; 340:409-17
Transfusion Requirements in
Critical Care
Randomized to 2 strategies
Liberal strategy:
– Maintain Hgb between 10-12
Restrictive strategy:
– Maintain Hgb between 7-9
Endpoints
– All cause mortality, MSOF
– Predefined subgroups: age > 55,
CAD, APACHE II > 20
Transfusion Requirements in
Critical Care
Restrictive Liberal
p
(n=418) (n=420)
ICU mortality 13.4%
16.2% 0.29
Death (30d)
18.7%
23.3%
0.11
ICU LOS
11.0
11.5
0.53
MODS
8.3
8.8
0.10
0.7%
2.9%
0.02
MI
Transfusion Requirements in
Critical Care
Patients with APACHE II < 20
Survival (%)
100
90
80
p=0.02
Restrictive
Liberal
70
60
50
0
5
10
15
Days
20
25
30
Transfusion Requirements in
Critical Care
Patients Younger than 55
Survival (%)
100
90
80
p=0.02
70
Restrictive
Liberal
60
50
0
5
10
15
Days
20
25
30
Transfusion Requirements in
Critical Care
 Conclusions
- Lower transfusion threshold was as
effective as higher trigger
- Lower threshold superior in some
subgroups
- Mechanism of worse outcomes with
liberal strategy unclear (? promotes
cytokine cascade, increased risk of
ARDS)
Transfusion Requirements in
Critical Care
 Editorial comment in NEJM
“This study has made it clear that a single
threshold for transfusion in all patients is
not appropriate…… With this knowledge,
more physicians will be able to follow the
dictum “first do no harm,” and we will have
a surplus of blood rather than a shortage.”
Ely et al. NEJM 1999: 340: 468.
Neuro Critical Care
Just one paper…
But may be very important !
Volume 61(8)
28 October 2003
pp 1047-1051
Detrimental effect of blood pressure reduction in the
first 24 hours of acute stroke onset
Oliveira-Filho, J. MD, PhD; Silva, S.C.S. MD; Trabuco, C.C. MD;
Pedreira, B.B. MD; Sousa, E.U. MD; Bacellar, A. MD
From the Department of Biomorphology (Dr. Oliveira-Filho), Health Sciences Institute, Federal University of Bahia;
the Neurology Service (Drs. Oliveira-Filho, Silva, Trabuco, Pedreira, and Bacellar), Hospital Sao Rafael, Monte
Tabor Foundation; and the Bahian School of Medicine and Public Health (Dr. Sousa), Salvador-Bahia, Brazil.
Results and Conclusions
The main result of the study was a strong,
independent relationship of BP course to
adverse outcome, with an almost twofold
increased risk of poor outcome for every 10%
decrease in the SBP over the first 24 hours.
The degree of SBP variation over the first 24
hours was unrelated to stroke severity but
robustly related to stroke outcome.
Oliveira-Filho J, Neurology 2003;61:1047
Results and Conclusions
The degree of SBP reduction was unrelated to
antihypertensive medication use (59% of patients), but
related to higher admission BP.
Even spontaneous decreases in BP may be harmful
to ischemic brain tissue. Cannot rule out that further
decreases with antihypertensive medications may
further worsen outcome.
Need for an acute intervention trial randomizing
patients for predefined BP values.
Oliveira-Filho J, Neurology 2003;61:1047
SUMMARY
NEW THERAPIES IN ICU PATIENTS (“Drivers”
of Mortality)
“Goal directed” early (ED) resuscitation
Rivers, NEJM 2001
30.5% mortality versus 46.5% (NNT = 6)
Intensive insulin therapy
Van den Berghe, NEJM 2001
10.6% mortality versus 20.2% (NNT = 20)
Low dose steroids in sepsis
Annane, JAMA 2002
53% mortality versus 63% (NNT = 10)
NEW THERAPIES (cont)
Low tidal volumes for ALI/ARDS
ARDS network, NEJM 2000
31.0% mortality versus 39.8% (NNT = 11)
“Appropriate” antibiotics
Kollef, Chest 1999
12.2% mortality versus 52.1% (NNT = 2.5)
“Invasive” strategy for VAP diagnosis
Fagon, Ann Int Med 2000
16.2% mortality versus 25.8% (NNT = 10)
NEW THERAPIES (cont)
Activate protein C (Xigris®) for sepsis
Bernard, NEJM 2001
24.7% mortality versus 30.8% (NNT = 16)
Intensivist led multidisciplinary ICU team
Young, Effective Clinical Practice 2000
Up to 60% reduction in mortality
LESS PROVEN THERAPIES
(no good placebo controls)
Transfusion practices
Pulmonary artery catheters
Have not been shown to improve
outcome
Liberal versus conservative fluid strategy
Prevent the “2nd hit” (DVT, VAP, GI
bleeding)
Acknowledgements
Marie R. Baldisseri, MD